Diagnostic accuracy of on-scan assessments compared to clinical assessments using a periodontal probe for detecting gingival recession: A cross-sectional study

Aim:
Compare the diagnostic accuracy of measuring gingival recession (GR) using a digital ruler on intraoral scans with clinical measurements using a UNC-15 probe.

Methods:
A prospective, observational, examiner-blinded, single-arm, cross-sectional, single-site study in adults aged 18+ with minimum of 20 teeth. All participants underwent an intraoral scan. Clinical GR was recorded on 6-sites (non-molar teeth) and 8-sites (molar teeth) by the principal examiner. Anonymised scans were assessed by the principal examiner twice (Ex1a, 1b) and two additional examiners (Ex2 and 3).

Results:
109 participants were assessed; all had at least one site of buccal GR. Ex1a on-scan measurements compared to clinical measurements showed 66.9% sensitivity and 92.5% specificity. There were considerable differences between central and proximal sites, central sites showed greater sensitivity 77.0% and specificity 81.4% compared to proximal sites 38.5% and 97.2% respectively. Diagnostic precision was robust, agreement for intra-examiner variability (Examiner-1a vs 1b) was 98.5% (95% CI 98.1%-98.8%). Inter-examiner variability was 78.9% (95% CI, 77.0%-80.6%) for Ex-1a-Ex-2 and 86.2% (95% CI, 84.8%-87.5%) for Ex-1a-Ex-3.

Conclusion:
Intraoral scan measurements are reproducible and can be utilised with accuracy. Measuring GR clinically is the current gold standard despite the limitations, intraoral scans can be utilised to overcome these limitations.